Post on 30-Aug-2021
Journal of Mechanics Engineering and Automation 4 (2014) 639-647
Performance Quality Testing under Combined Loading
of Polyethylene Pipes Reinforced with Aramid Fiber
Hilario Hernández-Moreno1, Erik Vargas-Rojas1, Víctor Manuel Sauce-Rangel1, Iván Mortera-Bravo2, Jorge Luis
González-Velázquez2, Maarten P. Kruijer3 and Margarita Navarrete-Montesinos4
1. ESIME (Escuela Superior de Ingeniería Mecánica y Eléctrica) Unidad Ticomán, Instituto Politécnico Nacional, Distrito Federal
07340, México
2. ESIQIE (Escuela Superior de Ingeniería Química e Industrias Extractivas), Instituto Politécnico Nacional, Distrito Federal 07300,
México
3. Department of Research & Development, Pipelife Netherlands B.V., Enkhuizen, P.O. Box 380; 1600 AJ, The Netherlands
4. Instituto de Ingeniería, Universidad Nacional Autónoma de México, Distrito Federal 04510, México
Received: May 19, 2014 / Accepted: June 05, 2014 / Published: August 25, 2014. Abstract: This paper presents the results of the performance quality testing of polyethylene pipes reinforced with aramid fibers, intended for applications such as discharging and gathering oil pipelines, and describes the test rig specifically designed for this purpose. The pipe specimens are submitted to impact with a device that simulates the collision of a pickaxe, and of a backhoe loader. After the impact, the pipes are tested under combined loading comprising internal pressure, and transverse loading; some pipe specimens without previous impact are tested as well. The results show that the reinforced thermoplastic pipes can fully withstand maximal operating pressure levels in the presence of damage and additional transverse loading. Key words: Combined pressure-bending testing, oil industry, reinforced thermoplastic pipe, composite materials.
1. Introduction
Discharge and gathering oil pipelines are feasible to
be affected by several forms of damage due to
corrosion, complex loading, falling rocks, terrain
movements and human activities like construction or
agriculture that could strike them and inducing indeed
mechanical damage. Metallic pipes are particularly
susceptible to these kinds of aggressions, so the oil
industry is looking into non-metallic solutions to cope
with these problems. In order to certify the capability
of non-metallic pipes, it is necessary to conduct a
series of mechanical tests that simulate these harsh
operating conditions. In this research, a non-metallic
composite pipe or RTP (reinforced thermoplastic pipe)
Corresponding author: Jorge Luis González-Velázquez,
Ph.D., research fields: mechanical metallurgy, fracture mechanics, corrosion, microstructural control and composites for oil industry. E-mail: jlgonzalezv@ipn.mx.
is made of an inner and an outer pipe fabricated with
HDPE (high density polyethylene), and an internal
tape that consists of aramid cords that are embedded,
but not impregnated by the HPDE [1].
Many studies related to thermoset composite pipes
[2-6] have been published considering combined loads,
as well as thermoplastic pipes [7]. However, there are
few works relaying on RTP under combined load
testing; in two of them, Kruijer et al. [8] investigated
polyethylene pipes reinforced with embedded aramid
fibers so as to identify its torsion-extension coupling,
and the viscoelastic behavior of steel reinforced
thermoplastic (polyethylene) pipe or s-RTP. With
regards to test methods, Alexander [9] develops a
series of full-scale tests on metallic repaired off-shore
raisers using composites comprising pressure,
pressure-tension, and pressure-bending tests, and
develops a combined four point flexural device in
DAVID PUBLISHING
D
Performance Quality Testing under Combined Loading of Polyethylene Pipes Reinforced with Aramid Fiber
640
which load is applied by means of two hydraulic rams.
Corona and Kyriakides [10] reported an experimental
procedure to evaluate buckling behavior under
external pressure and bending.
Regardless of the fact that there is an important
amount of research that has been conducted lately on
the study of thermoset-composite pipes, there are no
standardized tests intended for RTP that simulate
combined loading encountered in real service
conditions. This situation leads to evaluate the load
capacity of the new type of thermoplastic reinforced
pipes under harsh conditions encountered at on-shore
applications. So, in order to fulfill this requirement, a
series of tests are specifically developed to simulate
extreme operating conditions such as combined
flexural-pressure testing of undamaged specimens;
and flexural-pressure testing of pipes previously
stricken forcefully arising two kinds of damages: one
that simulates a pickaxe tool, and the impact produced
by a backhoe loader. The series of verification tests
are performed on pipe samples bent to a bending
radius of 1,500 mm.
The paper is organized as follows: Section 2
describes the test specimens under study; Section 3
presents the non-standardized tests considered;
Section 4 describes the testing setup in detail,
comprising the conditioning of the test specimens, the
impact testing devices, and the pressure-flexure
testing; Section 5 presents and analyzes the results of
the previous tests in terms of the measured parameters:
time, internal pressure, load applied onto the RTP, and
displacement during the flexure test; finally, Section 6
gives the conclusions.
2. Pipe Description
The aramid reinforced polyethylene pipes fabricated
by Soluforce have a nominal diameter of 127 mm
(5"), and are composed of four layers (Fig. 1). The
first layer is a thick polyethylene liner covered with
two embedded, but not-impregnated aramid (Twaron)
layers, each one composed of aramid rovings ducted
Fig. 1 Schematic cross section of a RTP pipe.
within polyethylene bands, with orientations of +55°
and -55° with respect to the pipe axis. The last layer is
a polyethylene protection layer treated with
UV-resistant additives.
3. Improved Acquisition
In order to evaluate the pipe load bearing capacity
to transport hydrocarbons safely, a series of
non-standardized tests that simulate harsh operating
conditions are proposed: three pipes receive an impact
equivalent to 14 J, simulating a pickaxe; three other
pipes are impacted with a 124 J strike, simulating a
backhoe loader. Then all the struck pipes are tested
under combined flexure-pressure to simulate adverse
conditions, and three others as in fabricated condition.
4. Test Setup Description
4.1 Specimens and Conditioning
The test specimens have a total length of 1,990 mm
and are machined at both ends according to manufacturer
specifications, so as to remove the aramid reinforcement
layers and the protective polyethylene layer leaving a
50 mm length of nude polyethylene liner at both ends,
so a couple of hermetic closures can be installed.
Conditioned specimens are shown in Fig. 2.
4.2 Impact Testing That Simulates a Pickaxe Tool Damage
In order to develop the impact testing, the
equivalent energy for the specific event is estimated to
4 mm
6.5 mm
4 mm
118 mm
Performance Quality Testing under Combined Loading of Polyethylene Pipes Reinforced with Aramid Fiber
641
Fig. 2 Specimens and conditioning: (a) raw specimen; (b) end closures and (c) conditioned specimen.
Fig. 3 Impact rig for simulating hand tool damage: (a) photograph and (b) schematic diagram.
be 14 J. The device used for this test is available at the
Engineering Institute of UNAM (Universidad Nacional
Autónoma de México), and it consists of a free fall
impact tester as shown in Fig. 3; it has a falling mass of
about 6.8 kg and an impact spherical head of 10 mm.
The calculated energy is measured with an
accelerometer connected to a data acquisition card and
a PC. Then, the necessary height to obtain a final
speed of 2 m/s at the moment of the impact can be
calculated; all specimens for this test are as in
fabricated condition except for their machined ends.
4.3 Impact Testing That Simulates a Backhoe Loader
Machine Damage
The experimental device for this test is an impact
test machine developed at IPN-ESIME (Instituto
Politécnico Nacional- Escuela Superior de Ingeniería
Mecánica y Eléctrica) Unidad Ticomán, which is also
conceived as a free fall impact tester. In order to
obtain the required energy for this impact, the height
for the test was obtained by trial and error using an
accelerometer and a data acquisition card connected to
a PC until the final speed of 4 m/s is reached. In Fig. 4,
a photo of the device and a schematic diagram are
presented.
4.4 Combined Loads Testing: Pressure and Flexure
In order to implement these tests, the experimental
setup consists of an universal mechanical testing
machine model Instron 8502, and a specifically
designed three point bending test device installed onto
the actuator head. The conditioned specimen with its
end closures, and the required hydraulic connections
are placed over the flexural rig (Fig. 5). Both the
specimen and the flexural rig are surrounded with a
transparent protective barrier supported with metallic
stiffeners. The flexural beam is a commercially
available C-shaped cross section aligned horizontally.
Pressurization is carried out using water by means
of two independent hydraulic circuits which operate
alternatively. The main circuit is under high pressure
and is used when the testing is running; the secondary
circuit is low pressure, and is used for replenishment
and drain. The high pressure is supplied by a
positive-displacement, reciprocating pump specifically
designed for this test, operated with a Shimadzu
electromechanical universal testing machine. The
hydraulic setup is depicted in Fig. 6.
The measured parameters are the flexure load, the
displacement of the actuator of the Instron testing
machine, and the specimen internal pressure. The
Instron machine control unit provides several analog
outputs; in this case two of these ports are used to
1,890 mm50 mm
1,280 mm
147
mm
131
mm
(a)
(b)
(c)
(a) (b)
Accelerometer
Guiding bar
Free falling bridge
with known mass
Calibrated height
Data acquisition card
Laptop
Test probe
Anvil
Impact
spherical
head
Performance Quality Testing under Combined Loading of Polyethylene Pipes Reinforced with Aramid Fiber
642
Fig. 4 Impact tester for simulating backhoe loader machine damage: (a) actual device and (b) diagram.
Fig. 5 Flexural setup: (a) lateral view and (b) cross section.
Fig. 6 Hydraulic circuits for combined pressure-flexure testing.
output the flexural load signal and the actuator
displacement signal. To obtain the pressure signal, a
voltage modulated signal transmitter is used. All
signals are acquired using a data acquisition card
connected to a notebook computer via a USB
(universal serial bus) port. In Fig. 7, the
instrumentation explicative diagram is presented.
To run the test, hydraulic pressure is increased up to
100 bar, which is the maximum operating pressure
specified by the manufacturer of the pipes, with a
pressure slope of 10 bar/min. Once the pressure is
reached and after verifying that hermeticity is assured,
the hydraulic actuator moves at 10 mm/min until a
bending radius of 1,500 mm of the specimen is
reached, equivalent to an actuator displacement equal
to 60 mm. Once the combined pressure and flexure
condition are obtained, they are sustained for not less
than 1 min, then the flexure load is released, and the
specimen is depressurized. During all this steps,
displacement, flexure load and pressure signals are
acquired at a rate of 1 Hz; this rate is judged
acceptable because the test is considered as in a
quasi-static condition. The ideal pressure and
displacement profile is depicted in Fig. 8, and
photographs of the actual test are shown in Fig. 9.
5. Results and Discussion
5.1 Impact
The impact test profiles shown in Fig. 10 consist of
an acceleration increment until an approximate value
of -7 m/s2; under frictionless and dragless ideal
conditions, this value should be the acceleration due to
Fig. 7 Instrumentation required for pressure-flexure testing.
(a) (b)
Pulleys
Lifting cable
Lifting
mechanism
Impact spherical head
Test probe
Anvil
Structure
Falling bridge
Accelerometer
Exchangeable
mass Laptop
Load-cell Instron 8502
Load adaptor
Test specimen
Flexural support
Anti-skid lateral bar
Support base
Flexural beam
Barrier base
Barrier metallic stiffener
Lateral barrier wall
Instron 8502 actuator
Load-cell Instron 8502
Load adaptor
Anti-skid lateral bar
Test specimen Lateral barrier wall
Barrier metallic
Base adaptor
Flexural support plate
Support base
Flexural beam
Instron 8502 actuator
Data acquisition
card
Pressure inlet (high pressure flexible hose)
Test probe
End closure End closure
Pressure transducerBall-type valve
Water feed and drain
stiffener
(a)
(b)
Data acquisition card
Pressure transmitter
Test probe with end closures
Pressure signal
Load-cell signal
Connections box
Load-cell signal
USB cable
Transmitter voltage
source input
Console
Instron
8502
Actuator
position
signal
Performance Quality Testing under Combined Loading of Polyethylene Pipes Reinforced with Aramid Fiber
643
Fig. 8 Ideal pressure and displacement profile.
Fig. 9 Combined flexure-pressure test: (a) test setup; (b) specimen to be tested; (c) instrumentation and (d) specimen under combined flexure and pressure condition.
gravity. After this acceleration limit, a rapid change in
acceleration is observed corresponding to the instant
when the specimen is impacted; finally a region of
several peaks is observed, corresponding to a series of
rebounds after the impact, however, they are neglected.
Thus, Figs. 10a-10c show the experimental graphs of
the 2 m/s impact test; in Figs. 11a-11c, the 4 m/s test
results are presented. The condensed results are shown
in Table 1 (speed at impact and kinetic energy).
5.2 Impact Damage Morphology
The damages produced in all pipes owing to the 2
m/s impact test show a spherical segment with an 8
mm diameter mark, as it can be observed in Fig. 12a.
The damage due to the impact at 4 m/s has also a
spherical mark of about 12 mm in diameter, but
Fig. 10 Specimen acceleration and speed profiles: (a) TA-04; (b) TA-05 and (c) TA-06.
presents some additional marks caused by the
rebounds after impact, as can be seen in Fig. 12b.
Afterward, during the combined pressure and flexure
loading, these marks tend to reduce their deformed
shape and to move closer towards the median surface
of the shell.
5.3 Combined Pressure-Flexure Results
The numerical treatment of the recorded data of
nine tests is done for each type of test: specimens
without damage, specimens previously impacted at 2
m/s, and specimens previously impacted at 4 m/s. The
Transverse load
Pressure
100 bar
(a) (b)
(c) (d)
2,000 N
Acc
eler
atio
n (m
/s2 )
Acc
eler
atio
n (m
/s2 )
Acc
eler
atio
n (m
/s2 )
Spee
d (m
/s)
Spee
d (m
/s)
Spee
d (m
/s)
Time (s) (b)
(a)
(c)Time (s)
Time (s)
Performance Quality Testing under Combined Loading of Polyethylene Pipes Reinforced with Aramid Fiber
644
Fig. 11 Specimen acceleration and speed profiles: (a) TA-07; (b) TA-08 and (c) TA-09.
final results are presented as graphs of pressure-time
in Fig. 13, load-time in Fig. 14, and load-displacement
Fig. 15. All the results corroborate that the specimens
behavior themselves as a whole structure or an end
product, showing geometry, dimensions, materials and
construction parameters, should be taken into account,
not only representing material properties.
5.3.1 Pressure as a Function of Time
In this case, pressure corresponds to an excitation
parameter. Experimental profiles of pressure-time
show a sawtooth profile, as expected after using a
reciprocating water pump to feed the system. The first
Table 1 Condensed impact test results.
Impact simulating hand tool damage
Specimen Impact speed (m/s) Maximum kinetic energy (J)
TA-04 1.9 12.2
TA-05 1.9 12.9
TA-06 1.9 12.2
Mean 1.9 12.4
Standard deviation 0.03 0.38 Coefficient of variation
0.015 0.030
Impact simulating backhoe loader machine
Specimen Impact speed (m/s) Maximum kinetic energy (J)
TA-07 3.7 46.5
TA-08 3.6 44.3
TA-09 3.6 45.1
Mean 3.7 45.3
Standard deviation 0.04 1.09 Coefficient of variation
0.012 0.024
(a) (b)
Fig. 12 Damage produced by impact: (a) 2 m/s and (b) 4 m/s.
segment corresponds to the increasing slope of the
pressure-time graph which is composed of two parts:
the first one is the beginning of a non-lineal
pressurization response, and the second is the lineal
ramp until the desired test pressure (100 bar); once it
is reached, the water pump supply is shutdown. From
this point on, a pressure decrease of about 10 bar is
observed due to small elongations on the pipe that
produce small increments in volume, a type of
behavior that can be attributed to a visco-elastic
response of the polyethylene. The last segment
corresponds to depressurization, with no remarks on
its behavior.
5.3.2 Load as a Function of Time
Load is a response parameter to both pressure and
Time (s)
Time (s)
Time (s)
Acc
eler
atio
n (m
/s2 )
Acc
eler
atio
n (m
/s2 )
Acc
eler
atio
n (m
/s2 )
Spee
d (m
/s)
Spee
d (m
/s)
Spee
d (m
/s)
(c)
(a)
(b)
Performance Quality Testing under Combined Loading of Polyethylene Pipes Reinforced with Aramid Fiber
645
Fig. 13 Pressure profiles: (a) specimens without damage; (b) specimens impacted at 2 m/s and (c) specimens impacted at 4 m/s.
the desired flexion displacement. In Fig. 14, an
increase of the load can be observed at the beginning
of the test because of a circumferential restrained
expansion, due to the hydraulic pressure increase.
Subsequently, a load augmentation is registered
corresponding to the displacement of the actuator
acting against the already pressurized bending pipe
until the equivalent displacement of 1,500 mm bending
radius is achieved with the three-point-bending test.
On the top of the curve, there is a significant reduction
Fig. 14 Load profiles: (a) specimens without damage; (b) specimens impacted at 2 m/s and (c) specimens impacted at 4 m/s.
of the load response due to the displacements within
the pipe because of its visco-elastic behavior, typical
of thermoplastic materials. The last segment
corresponds to the flexural unloading.
Analyzing the graph load-time, the effect of the
damages produced during impact test is imperceptible.
Condensed values of the actual maximum pressure
and load are presented in Table 2.
5.3.3 Load as a Function of Displacement
In Fig. 15, the load-displacement plot shows a
Time (s)
Time (s)
Time (s)
Pre
ssur
e (b
ar)
Pre
ssur
e (b
ar)
Pre
ssur
e (b
ar)
0 200 400 600 800 1,000 1,200 1,400 1,600
0 200 400 600 800 1,000 1,200 1,400 1,600
0 200 400 600 800 1,000 1,200 1,400 1,600 0 200 400 600 800 1,000 1,200 1,400 1,600
0 200 400 600 800 1,000 1,200 1,400 1,600
0 200 400 600 800 1,000 1,200 1,400 1,600
Time (s)
Time (s)
Time (s)
0
5,000
10,000
15,000
20,000
25,000
0
5,000
10,00010,000
15,000
20,000
25,000
0
5,000
10,000
15,000
20,000
25,000
Loa
d (
N)
Loa
d (
N)
Loa
d (
N)
Performance Quality Testing under Combined Loading of Polyethylene Pipes Reinforced with Aramid Fiber
646
Table 2 Maximum pressure and load.
Specimen Maximum pressure (bar) Maximum load (N)
TA-01-bis 100.2 21,601.3
TA-02-bis 99.1 21,471.3
TA-03 101.8 21,063.7
TA-04 102.8 20,552.4
TA-05 100.3 19,084.5
TA-06 102.2 18,839.8
TA-07 100.3 22,020.5
TA-08 101.7 22,020.5
TA-09 102.3 21,531.2
Fig. 15 Load versus displacement diagram: (a) specimens without damage; (b) specimens with a 2 m/s impact and (c) specimens with a 4 m/s impact.
non-linear behavior followed by a permanent
elongation after unloading; this behavior may be
attributed to a visco-elastic response combined with
plasticity. The plastic strain is in the order of
magnitude of 25 mm for specimens without damage,
and of 30 mm for specimens previously impacted.
6. Conclusions
Based on experimental observations, it can be stated
that the tested specimens withstand the maximal
operating pressure specified by the manufacturer (100
bar), as well as additional combined flexure load in
the presence of damage, in this case as generated by a
series of impacts that simulate either a pickaxe tool or
a backhoe loader.
These tests show a global behavior of the product
(the pipe as a whole structure), thus not representing
only material properties.
The performed tests are complementary to those
specified either on international standards or on local
regulations/norms, and they can offer a better
approach for design and analysis of pipes submitted to
harsh conditions.
Acknowledgments
The authors gratefully thank IPN (Instituto
Politécnico Nacional) faculty member Antonio
Mosqueda Sánchez, as well as students Sergio Ávila
Hernández, Renato Alcázar Morales, Diego Camarena
Arellano, Daniel Hernández M., Alexander Morales
Gómez and Honorio Ortiz Hernández for their
collaboration and support.
References
[1] Kruijer, M. P., Warnet, L. L., and Akkerman, R. 2005. “Analysis of the Mechanical Properties of a RTP (Reinforced Thermoplastic Pipe).” Composites: Part A 36: 291-300.
[2] El-Assal, A. M., and Khashaba, U. A. 2007. “Fatigue Analysis of Unidirectional GFRP Composites under Combined Bending and Torsional Loads.” Composite Structures 79: 599-605.
[3] Hur, S. H., Son, H. J., Kweon, J. H., and Choi, J. H. 2008.
Displacement (mm)
Displacement (mm)
Displacement (mm)
0
5,000
10,000
15,000
20,000
25,000
0
0
5,000
5,000
10,000
10,000
15,000
15,000
20,000
20,000
25,000
25,000
Loa
d (
N)
Loa
d (
N)
Loa
d (
N)
Performance Quality Testing under Combined Loading of Polyethylene Pipes Reinforced with Aramid Fiber
647
“Postbuckling of composite cylinders under external hydrostatic pressure.” Composite Structures 86: 114-24.
[4] Guedes, R. M. 2009. “Stress-Strain Analysis of a Cylindrical Pipe Subjected to a Transverse Load and Large Deflections.” Composite Structures 88: 188-94.
[5] Lam, K. Y., Zong, Z., and Wang, Q. X. 2003. “Dynamic Response of a Laminated Pipeline on the Seabed Subjected to Underwater Shock.” Composites Part B 34: 59-66.
[6] Weaver, P. M. 2003. “The Effect of Extension/Twist Anisotropy on Compression Buckling in Cylindrical Shells.” Composites Part B 34: 251-60.
[7] Krishnaswamy, R. K. 2005. “Analysis of Ductile and Brittle Failures from Creep Rupture Testing of HDPE
(High-Density Polyethylene) Pipes.” Polymer 46: 11664-72.
[8] Kruijer, M. P., Warnet, L. L., and Akkerman, R. 2006. “Modeling of the Viscoelastic Behaviour of Steel Reinforced Thermoplastic Pipes.” Composites Part A 37: 356-67.
[9] Alexander, C. 2007. “Evaluating the Use of Composite Materials in Reinforcing Offshore Risers Using Full-Scale Testing Methods.” Presented at the Conference ASME International Offshore Pipeline Forum, Texas, USA.
[10] Corona, E., and Kyriakides, S. 2000. “Asymmetric Collapse Modes of Pipes under Combined Bending and External Pressure.” Journal of Engineering and Materials 126: 1232-39.